Wheat Dataset Research Articles

A linear profit function for economic weights of linear phenotypic selection indices in plant breeding

AbstractThe profit function (net returns minus costs) allows breeders to derive trait economic weights to predict the net genetic merit (H) using the linear phenotypic selection index (LPSI). Economic weight is the increase in profit achieved by improving a particular trait by one unit and should reflect the market situation and not only preferences or arbitrary values. In maize (Zea mays L.) and wheat (Triticum aestivum) breeding programs, only grain yield has a specific market price, which makes application of a profit function difficult. Assuming the traits’ phenotypic values have multivariate normal distribution, we used the market price of grain yield and its conditional expectation given all the traits of interest to construct a profit function and derive trait economic weights in maize and wheat breeding. Using simulated and real maize and wheat datasets, we validated the profit function by comparing its results with the results obtained from a set of economic weights from the literature. The criteria to validate the function were the estimated values of the LPSI selection response and the correlation between LPSI and H. For our approach, the maize and wheat selection responses were 1,567.13 and 1,291.5, whereas the correlations were .87 and .85, respectively. For the other economic weights, the selection responses were 0.79 and 2.67, whereas the correlations were .58 and .82, respectively. The simulated dataset results were similar. Thus, the profit function is a good option to assign economic weights in plant breeding.

An Optimized Technique for RNA Prediction Based on Neural Network

Pathway reconstruction, which remains a primary goal for many investigations, requires accurate inference of gene interactions and causality. Non-coding RNA (ncRNA) is studied because it has a significant regulatory role in many plant and animal life activities, but interacting micro-RNA (miRNA) and long non-coding RNA (lncRNA) are more important. Their interactions not only aid in the in-depth research of genes’ biological roles, but also bring new ideas for illness detection and therapy, as well as plant genetic breeding. Biological investigations and classical machine learning methods are now used to predict miRNA-lncRNA interactions. Because biological identification is expensive and time-consuming, machine learning requires too much manual intervention, and the feature extraction process is difficult. This research presents a deep learning model that combines the advantages of convolutional neural networks (CNN) and bidirectional long short-term memory networks (Bi-LSTM). It not only takes into account the connection of information between sequences and incorporates contextual data, but it also thoroughly extracts the sequence data’s features. On the corn data set, cross-checking is used to evaluate the model’s performance, and it is compared to classical machine learning. To acquire a superior classification effect, the proposed strategy was compared to a single model. Additionally, the potato and wheat data sets were utilized to evaluate the model, with accuracy rates of 95% and 93%, respectively, indicating that the model had strong generalization capacity.

RNA-Seq Data Processing in Plant-Pathogen Interaction System: A Case Study.

In RNA-seq data processing, short reads are usually aligned from one species against its own genome sequence; however, in plant-pathogen interaction systems, reads from both host and pathogen samples are blended together. In contrast with single-genome analyses, both pathogen and host reference genomes are involved in the alignment process. In such circumstances, the order in which the alignment is carried out, whether the host or pathogen is aligned first, or if both genomes are aligned simultaneously, influences the read counts of certain genes. This is a problem, especially at advanced infection stages. It is crucial to have an appropriate strategy for aligning the reads to their respective genomes, yet the existing strategies of either sequential or parallel alignment become problematic when mapping mixed reads to their corresponding reference genomes. The challenge lies in the determination of which reads belong to which species, especially when homology exists between the host and pathogen genomes. This chapter proposes a combo-genome alignment strategy, which was compared with existing alignment scenarios. Simulation results demonstrated that the degree of discrepancy in the results is correlated with phylogenetic distance of the two species in the mixture which was attributable to the extent of homology between the two genomes involved. This correlation was also found in the analysis using two real RNA-seq datasets of Fusarium-challenged wheat plants. Comparisons of the three RNA-seq processing strategies on three simulation datasets and two real Fusarium-infected wheat datasets showed that an alignment to a combo-genome, consisting of both host and pathogen genomes, improves mapping quality as compared to sequential alignment procedures.

Improved image recognition via Synthetic Plants using 3D Modelling with Stochastic Variations

This research extends previous plant modelling using L-systems by means of a novel arrangement comprising synthetic plants and a refined global wheat dataset in combination with a synthetic inference application. The study demonstrates an application with direct recognition of real plant stereotypes, and augmentation via a plant-wide stochastic growth variation structure. The study showed that the automatic annotation and counting of wheat heads using the Global Wheat dataset images provides a time and cost saving over traditional manual approaches and neural networks. This study introduces a novel synthetic inference application using a plant-wide stochastic variation system, resulting in improved structural dataset hierarchy. The research demonstrates a significantly improved L-system that can more effectively and more accurately define and distinguish wheat crop characteristics.

Comprehensive, integrative genomic analysis of microRNA expression profiles in different tissues of two wheat cultivars with different traits.

Wheat is one of the most important food sources on Earth. MicroRNAs (miRNA) play important roles in wheat productivity. To identify wheat miRNAs, we constructed and sequenced sRNA libraries from leaves and roots of two wheat cultivars (RAC875 and Kukri) with many different traits. Given that available miRNA wheat complement in the plant-specific database PmiREN ( https://pmiren.com ) does not include root tissues and root-associated miRNAs might thus be missing, we performed first the prediction of novel miRNAs using the sRNAbench tool. We found a total of 150 putatively novel miRNA genes with expression of both arms from 289 unique mature sequences and nearly 30% of all miRNA reads in roots corresponded to novel miRNAs. In contrast, this figure in leaves dropped to under 3%, confirming the undersampling of roots in the complement of known miRNAs. By using 120 publicly available wheat datasets, 598 Zea mays small RNA libraries, 64 plant species genomes, wheat degradome library, and functional enrichment analysis, a subset of novel miRNAs were confirmed as bona-fide miRNAs. Of the total 605 miRNAs identified in this study inclusive of 316 known miRNAs, 528 miRNAs were shared by both cultivars, 429 miRNAs were shared by both root tissues and 329 miRNAs were shared by both leaf tissues. In addition, 32 miRNAs were specific to Kukri while 45 miRNAs were specific to RAC875. These miRNAs had diverse functions, such as regulation of gene transcription, protein translation, energy metabolism, and cell cycle progression. Our data provide a genome-wide miRNA expression profile in these two wheat cultivars and help functional studies of wheat genomics.

Mega-scale Bayesian regression methods for genome-wide prediction and association studies with thousands of traits.

Large-scale phenotype data are expected to increase the accuracy of genome-wide prediction and the power of genome-wide association analyses. However, genomic analyses of high-dimensional, highly correlated traits are challenging. We developed a method for implementing high-dimensional Bayesian multivariate regression to simultaneously analyze genetic variants underlying thousands of traits. As a demonstration, we implemented the BayesC prior in the R package MegaLMM. Applied to Genomic Prediction, MegaBayesC effectively integrated hyperspectral reflectance data from 620 hyperspectral wavelengths to improve the accuracy of genetic value prediction on grain yield in a wheat dataset. Applied to Genome-Wide Association Studies, we used simulations to show that MegaBayesC can accurately estimate the effect sizes of QTL across a range of genetic architectures and causes of correlations among traits. To apply MegaBayesC to a realistic scenario involving whole-genome marker data, we developed a 2-stage procedure involving a preliminary step of candidate marker selection prior to multivariate regression. We then used MegaBayesC to identify genetic associations with flowering time in Arabidopsis thaliana, leveraging expression data from 20,843 genes. MegaBayesC selected 15 single nucleotide polymorphisms as important for flowering time, with 13 located within 100 kb of known flowering-time related genes, a higher validation rate than achieved by a single-stage analysis using only the flowering time data itself. These results demonstrate that MegaBayesC can efficiently and effectively leverage high-dimensional phenotypes in genetic analyses.

A Comparison between Three Tuning Strategies for Gaussian Kernels in the Context of Univariate Genomic Prediction.

Genomic prediction is revolutionizing plant breeding since candidate genotypes can be selected without the need to measure their trait in the field. When a reference population contains both phenotypic and genotypic information, it is trained by a statistical machine learning method that is subsequently used for making predictions of breeding or phenotypic values of candidate genotypes that were only genotyped. Nevertheless, the successful implementation of the genomic selection (GS) methodology depends on many factors. One key factor is the type of statistical machine learning method used since some are unable to capture nonlinear patterns available in the data. While kernel methods are powerful statistical machine learning algorithms that capture complex nonlinear patterns in the data, their successful implementation strongly depends on the careful tuning process of the involved hyperparameters. As such, in this paper we compare three methods of tuning (manual tuning, grid search, and Bayesian optimization) for the Gaussian kernel under a Bayesian best linear unbiased predictor model. We used six real datasets of wheat (Triticum aestivum L.) to compare the three strategies of tuning. We found that if we want to obtain the major benefits of using Gaussian kernels, it is very important to perform a careful tuning process. The best prediction performance was observed when the tuning process was performed with grid search and Bayesian optimization. However, we did not observe relevant differences between the grid search and Bayesian optimization approach. The observed gains in terms of prediction performance were between 2.1% and 27.8% across the six datasets under study.

Adaptive Artificial Neural Network in near infrared spectroscopy for standard-free calibration transfer

Near infrared spectroscopy (NIRS) is a secondary analytical method that could use machine learning to predict components of interest for various applications. Calibration transfer (CT) algorithms have been used to transfer the existing NIRS calibration model to another NIRS instruments to reduce the involvement of expensive and tedious conventional chemical analytical methods. However, most of CT algorithms involve both primary and secondary instruments to acquire NIRS signals from standard samples for spectral standardization. On the other hand, artificial neural network (ANN) that can adapt to new environmental conditions may be an alternative standard-free CT algorithm in transferring a trained ANN to secondary instruments without the involvement of the primary instrument. Although ANN has been widely implemented in NIRS research as a calibration model, its feasibility as an alternative CT method has not been evaluated. Thus, this study evaluates the feasibility of an adaptive ANN (AANN) in transferring ANN model from primary to secondary instruments using two traceable NIRS datasets. First, ANN and ANN with principal components (PCs-ANN) were developed and optimized using Bayesian learning algorithm. After that, these ANNs were adapted to secondary instruments using respective transfer data, in which, only the weights and biases of the ANNs were updated. Findings show that the lowest averaged RMSEP was achieved by the proposed PCs-AANN and AANN in the corn and wheat datasets, respectively. Particularly, the computation cost of having a calibration model in secondary instruments has been substantially reduced by means of the proposed AANN algorithm.

Water as a Probe for Standardization of Near-Infrared Spectra by Mutual-Individual Factor Analysis.

The standardization of near-infrared (NIR) spectra is essential in practical applications, because various instruments are generally employed. However, standardization is challenging due to numerous perturbations, such as the instruments, testing environments, and sample compositions. In order to explain the spectral changes caused by the various perturbations, a two-step standardization technique was presented in this work called mutual–individual factor analysis (MIFA). Taking advantage of the sensitivity of a water probe to perturbations, the spectral information from a water spectral region was gradually divided into mutual and individual parts. With aquaphotomics expertise, it can be found that the mutual part described the overall spectral features among instruments, whereas the individual part depicted the difference of component structural changes in the sample caused by operation and the measurement conditions. Furthermore, the spectral difference was adjusted by the coefficients in both parts. The effectiveness of the method was assessed by using two NIR datasets of corn and wheat, respectively. The results showed that the standardized spectra can be successfully predicted by using the partial least squares (PLS) models developed with the spectra from the reference instrument. Consequently, the MIFA offers a viable solution to standardize the spectra obtained from several instruments when measurements are affected by multiple factors.

Spectral Encoder to Extract the Features of Near-Infrared Spectra for Multivariate Calibration.

An autoencoder architecture was adopted for near-infrared (NIR) spectral analysis by extracting the common features in the spectra. Three autoencoder-based networks with different purposes were constructed. First, a spectral encoder was established by training the network with a set of spectra as the input. The features of the spectra can be encoded by the nodes in the bottleneck layer, which in turn can be used to build a sparse and robust model. Second, taking the spectra of one instrument as the input and that of another instrument as the reference output, the common features in both spectra can be obtained in the bottleneck layer. Therefore, in the prediction step, the spectral features of the second can be predicted by taking the reverse of the decoder as the encoder. Furthermore, transfer learning was used to build the model for the spectra of more instruments by fine-tuning the trained network. NIR datasets of plant, wheat, and pharmaceutical tablets measured on multiple instruments were used to test the method. The multi-linear regression (MLR) model with the encoded features was found to have a similar or slightly better performance in prediction compared with the partial least-squares (PLS) model.

A hierarchical Bayesian approach to dynamic ordinary differential equations modeling for repeated measures data on wheat growth

Experimental data collected on growth and development of plants over a growing season are typically analyzed using a linear mixed model, analogous to a hierarchical linear model in a Bayesian setting. Alternative modeling approaches for repeated measures data involve non-linear models such as logistic regression and ecophysiological dynamic models based on a system of ordinary differential equations (ODE). Yet, current implementations of ODE models are mostly deterministic in nature, which negates recognition of uncertainty in the data generation process and thus impairs inference and prediction. The primary objective of this study was to demonstrate the use of a dynamic ODE model within a Bayesian framework to make stochastic inference on system-level parameters. A secondary objective was to compare the predictive performance of an ODE model relative to methodologies more commonly used for repeated measures data from designed experiments, namely hierarchical linear models and hierarchical non-linear models. Using a hierarchical Bayesian implementation, we fit all three types of models to data on leaf area index (LAI) and biomass from a winter wheat dataset. In the context of this application, none of the modeling approaches clearly outperformed any other in terms of goodness of fit or prediction accuracy as indicated by similar posterior median values for root mean squared error (RMSE), Willmott’s agreement index (d), and Nash-Sutcliffe efficiency (NSE). The prediction statistics for the ODE, linear, and non-linear models respectively, were: RMSEp of 1.38, 1.14, and 1.19; dp of 0.91, 0.93, and 0.93; and NSEp of 0.69, 0.78, and 0.76 for LAI and RMSEp of 274.08, 253.04, and 207.63; dp of 0.95, 0.95, and 0.97; and NSEp of 0.82, 0.84, and 0.89 for biomass. The dynamic ODE model enabled biologically meaningful system-level inferences relevant to the research questions that were not possible when using the hierarchical linear or non-linear modeling approaches.

A General-Purpose Machine Learning R Library for Sparse Kernels Methods With an Application for Genome-Based Prediction.

The adoption of machine learning frameworks in areas beyond computer science have been facilitated by the development of user-friendly software tools that do not require an advanced understanding of computer programming. In this paper, we present a new package (sparse kernel methods, SKM) software developed in R language for implementing six (generalized boosted machines, generalized linear models, support vector machines, random forest, Bayesian regression models and deep neural networks) of the most popular supervised machine learning algorithms with the optional use of sparse kernels. The SKM focuses on user simplicity, as it does not try to include all the available machine learning algorithms, but rather the most important aspects of these six algorithms in an easy-to-understand format. Another relevant contribution of this package is a function for the computation of seven different kernels. These are Linear, Polynomial, Sigmoid, Gaussian, Exponential, Arc-Cosine 1 and Arc-Cosine L (with L = 2, 3, … ) and their sparse versions, which allow users to create kernel machines without modifying the statistical machine learning algorithm. It is important to point out that the main contribution of our package resides in the functionality for the computation of the sparse version of seven basic kernels, which is indispensable for reducing computational resources to implement kernel machine learning methods without a significant loss in prediction performance. Performance of the SKM is evaluated in a genome-based prediction framework using both a maize and wheat data set. As such, the use of this package is not restricted to genome prediction problems, and can be used in many different applications.

Leaf pigment retrieval using the PROSAIL model: Influence of uncertainty in prior canopy-structure information

Leaf pigments are critical indicators of plant photosynthesis, stress, and physiological conditions. Inversion of radiative transfer models (RTMs) is a promising method for robustly retrieving leaf biochemical traits from canopy observations, and adding prior information has been effective in alleviating the “ill-posed” problem, a major challenge in model inversion. Canopy structure parameters, such as leaf area index (LAI) and average leaf inclination angle (ALA), can serve as prior information for leaf pigment retrieval. Using canopy spectra simulated from the PROSAIL model, we estimated the effects of uncertainty in LAI and ALA used as prior information for lookup table-based inversions of leaf chlorophyll (Cab) and carotenoid (Car). The retrieval accuracies of the two pigments were increased by use of the priors of LAI (RMSE of Cab from 7.67 to 6.32 μg cm−2, Car from 2.41 to 2.28 μg cm−2) and ALA (RMSE of Cab from 7.67 to 5.72 μg cm−2, Car from 2.41 to 2.23 μg cm−2). However, this improvement deteriorated with an increase of additive and multiplicative uncertainties, and when 40% and 20% noise was added to LAI and ALA respectively, these priors ceased to increase retrieval accuracy. Validation using an experimental winter wheat dataset also showed that compared with Car, the estimation accuracy of Cab increased more or deteriorated less with uncertainty in prior canopy structure. This study demonstrates possible limitations of using prior information in RTM inversions for retrieval of leaf biochemistry, when large uncertainties are present.

Comparing gradient boosting machine and Bayesian threshold BLUP for genome-based prediction of categorical traits in wheat breeding.

Genomic selection (GS) is a predictive methodology that is changing plant breeding. Genomic selection trains a statistical machine-learning model using available phenotypic and genotypic data with which predictions are performed for individuals that were only genotyped. For this reason, some statistical machine-learning methods are being implemented in GS, but in order to improve the selection of new genotypes early in the prediction process, the exploration of new statistical machine-learning algorithms must continue. In this paper, we performed a benchmarking study between the Bayesian threshold genomic best linear unbiased predictor model (TGBLUP; popular in GS) and the gradient boosting machine (GBM). This comparison was done using four real wheat (Triticum aestivum L.) data sets with categorical traits measured in terms of two metrics: the proportion of cases correctly classified (PCCC) and the Kappa coefficient in the testing set. Under 10 random partitions with four different sizes of testing proportions (20, 40, 60, and 80%), we compared the two algorithms and found that in three of the four data sets, the GBM outperformed the TGBLUP model in terms of both metrics (PCCC and Kappa coefficient). In the larger data sets (Data Sets 3 and 4), the gain in terms of prediction accuracy of the GBM was considerably significant. For this reason, we encourage more research using the GBM in GS to evaluate its virtues in terms of prediction performance in the context of GS.

Outdoor Plant Segmentation With Deep Learning for High-Throughput Field Phenotyping on a Diverse Wheat Dataset.

Robust and automated segmentation of leaves and other backgrounds is a core prerequisite of most approaches in high-throughput field phenotyping. So far, the possibilities of deep learning approaches for this purpose have not been explored adequately, partly due to a lack of publicly available, appropriate datasets. This study presents a workflow based on DeepLab v3+ and on a diverse annotated dataset of 190 RGB (350 x 350 pixels) images. Images of winter wheat plants of 76 different genotypes and developmental stages have been acquired throughout multiple years at high resolution in outdoor conditions using nadir view, encompassing a wide range of imaging conditions. Inconsistencies of human annotators in complex images have been quantified, and metadata information of camera settings has been included. The proposed approach achieves an intersection over union (IoU) of 0.77 and 0.90 for plants and soil, respectively. This outperforms the benchmarked machine learning methods which use Support Vector Classifier and/or Random Forrest. The results show that a small but carefully chosen and annotated set of images can provide a good basis for a powerful segmentation pipeline. Compared to earlier methods based on machine learning, the proposed method achieves better performance on the selected dataset in spite of using a deep learning approach with limited data. Increasing the amount of publicly available data with high human agreement on annotations and further development of deep neural network architectures will provide high potential for robust field-based plant segmentation in the near future. This, in turn, will be a cornerstone of data-driven improvement in crop breeding and agricultural practices of global benefit.

Mark-Spectra: A convolutional neural network for quantitative spectral analysis overcoming spatial relationships

Spectral analysis is one of the most important and widely used methods for chemometrics in the field of agriculture, and convolutional neural network (CNN) models have achieved excellent performance on spectral analysis. The critical drawback of the CNN approach is that it preserves the spatial relationships among adjacent wavelengths, which contribute to collinearity and redundancies rather than relevant effective information. To confirm this observation, the distribution of characteristic wavelengths extracted by different methods (include F-test, importance weights, and CNN) are visualized in this paper. A convolutional neural network for quantitative spectral analysis, named Mark-Spectra, is presented to overcome spatial relationships and to improve the model performance. A layer (Mark layer) is introduced as part of Mark-Spectra, which is used to overcome spatial relationship of raw spectral data. Mark-Spectra model is compared with three CNN models using three open accessed visible and near infrared spectroscopic datasets (corn, wheat and soil). Mark-Spectra model outperforms the other three convolutional neural network models on two datasets (except dataset of wheat, due to lesser number of features), and it cost much less training time than the others. In addition, this paper compares Mark-Spectra with two classical neural network-based algorithms, principal component analysis – artificial neural network (PCA-ANN) and extreme learning machine (ELM). Mark-Spectra performed best in soil dataset, and ELM performed best in corn and wheat datasets, respectively. These results can illustrate that Mark-Spectra is still limited with the characteristic of raw spectral data (e.g., the number of samples and features), which is a fundamental fact of deep learning-based methods, but it performed better than the other CNN models and reduced the dependence of sample size due to overcoming spatial relationships.

An interpretable deep learning approach for calibration transfer among multiple near-infrared instruments

Spectroscopic techniques have been widely applied in agricultural applications. The development of calibration transfer is promising for the robust analysis of spectral data collected by varying instruments. The reliance on standard samples for standardization approaches remains a critical challenge for on-site applications. In this study, a deep learning approach, named DeepTranSpectra (DTS), is proposed to transfer convolutional neural network models among multiple near-infrared spectrometers with different types. The proposed DTS approach effectively avoids the requirements for standard samples by using labeled samples of slave instruments. The calibration transfer analysis is investigated on a soybean meal and one wheat dataset for predicting moisture and crude protein contents. The developed DTS approach demonstrates improved transfer performance compared with three popular standardization approaches, including piecewise direct standardization (PDS), canonical correlation analysis (CCA), and slope and bias correction (SBC). A feature visualization method is leveraged to interpret the transfer mechanism of the DTS approach. The interpretation results show that the DTS approach refines the model parameters to adapt to slave devices based on critical features of the master calibration. The DTS approach provides advanced reliability under different sample selections in Monte-Carlo cross-validation. The integration of deep learning approaches with calibration transfer analysis facilitates agricultural applications for emerging deep learning-based chemometric analysis.

Genome-enabled prediction for sparse testing in multi-environmental wheat trials.

Sparse testing in genome-enabled prediction in plant breeding can be emulated throughout different line allocations where some lines are observed in all environments (overlap) and others are observed in only one environment (nonoverlap). We studied three general cases of the composition of the sparse testing allocation design for genome-enabled prediction of wheat (Triticum aestivum L.) breeding: (a) completely nonoverlapping wheat lines in environments, (b) completely overlapping wheat lines in all environments, and (c) a proportion of nonoverlapping/overlapping wheat lines allocated in the environments. We also studied several cases in which the size of the testing population was systematically decreased. The study used three extensive wheat data sets (W1, W2, and W3). Three different genome-enabled prediction models (M1-M3) were used to study the effect of the sparse testing in terms of the genomic prediction accuracy. Model M1 included only main effects of environments and lines; M2 included main effects of environments, lines, and genomic effects; whereas the remaining model (M3) also incorporated the genomic × environment interaction (GE). The results show that the GE component of the genome-based model M3 captures a larger genetic variability than the main genomic effects term from models M1 and M2. In addition, model M3 provides higher prediction accuracy than models M1 and M2 for the same allocation designs (different combinations of nonoverlapping/overlapping lines in environments and training set sizes). Overlapped sets of 30-50 lines in all the environments provided stable genomic-enabled prediction accuracy. Reducing the size of the testing populations under all allocation designs decreases the prediction accuracy, which recovers when more lines are tested in all environments. Model M3 offers the possibility of maintaining the prediction accuracy throughout both extreme situations of all nonoverlapping lines and all overlapping lines.

A Deep Learning-Based Method for Automatic Assessment of Stomatal Index in Wheat Microscopic Images of Leaf Epidermis.

The stomatal index of the leaf is the ratio of the number of stomata to the total number of stomata and epidermal cells. Comparing with the stomatal density, the stomatal index is relatively constant in environmental conditions and the age of the leaf and, therefore, of diagnostic characteristics for a given genotype or species. Traditional assessment methods involve manual counting of the number of stomata and epidermal cells in microphotographs, which is labor-intensive and time-consuming. Although several automatic measurement algorithms of stomatal density have been proposed, no stomatal index pipelines are currently available. The main aim of this research is to develop an automated stomatal index measurement pipeline. The proposed method employed Faster regions with convolutional neural networks (R-CNN) and U-Net and image-processing techniques to count stomata and epidermal cells, and subsequently calculate the stomatal index. To improve the labeling speed, a semi-automatic strategy was employed for epidermal cell annotation in each micrograph. Benchmarking the pipeline on 1,000 microscopic images of leaf epidermis in the wheat dataset (Triticum aestivum L.), the average counting accuracies of 98.03 and 95.03% for stomata and epidermal cells, respectively, and the final measurement accuracy of the stomatal index of 95.35% was achieved. R2 values between automatic and manual measurement of stomata, epidermal cells, and stomatal index were 0.995, 0.983, and 0.895, respectively. The average running time (ART) for the entire pipeline could be as short as 0.32 s per microphotograph. The proposed pipeline also achieved a good transferability on the other families of the plant using transfer learning, with the mean counting accuracies of 94.36 and 91.13% for stomata and epidermal cells and the stomatal index accuracy of 89.38% in seven families of the plant. The pipeline is an automatic, rapid, and accurate tool for the stomatal index measurement, enabling high-throughput phenotyping, and facilitating further understanding of the stomatal and epidermal development for the plant physiology community. To the best of our knowledge, this is the first deep learning-based microphotograph analysis pipeline for stomatal index assessment.

Multi-trait genomic-enabled prediction enhances accuracy in multi-year wheat breeding trials.

Implementing genomic-based prediction models in genomic selection requires an understanding of the measures for evaluating prediction accuracy from different models and methods using multi-trait data. In this study, we compared prediction accuracy using six large multi-trait wheat data sets (quality and grain yield). The data were used to predict 1 year (testing) from the previous year (training) to assess prediction accuracy using four different prediction models. The results indicated that the conventional Pearson’s correlation between observed and predicted values underestimated the true correlation value, whereas the corrected Pearson’s correlation calculated by fitting a bivariate model was higher than the division of the Pearson’s correlation by the squared root of the heritability across traits, by 2.53–11.46%. Across the datasets, the corrected Pearson’s correlation was higher than the uncorrected by 5.80–14.01%. Overall, we found that for grain yield the prediction performance was highest using a multi-trait compared to a single-trait model. The higher the absolute genetic correlation between traits the greater the benefits of multi-trait models for increasing the genomic-enabled prediction accuracy of traits.